1. Technical Field of the Invention
The present invention relates in general to the telecommunications field and, in particular, to a method for maintaining the reliable operation of a distributed network of service nodes overlaying a telecommunications network.
2. Description of Related Art
A basic customer requirement that operators of telecommunications networks must meet is that the networks operate reliably. In particular, the network requirements imposed on basic switching functions and basic subscriber services are extremely high. When more advanced services are introduced into the conventional networks, they usually adhere to the same high operational requirements even if the new advanced services are not as important or critical as the basic services already being provided.
Increased deregulation and competition have caused a recent trend in the telecommunications field, wherein new service providers are basing their advanced services on the existing infrastructures of basic switching functions and subscriber services being provided by the conventional network operators. Consequently, overlayed networks of service nodes are being created, which are characterized as generic equipment provided by a plurality of suppliers. Recently, a great deal of resources have been expended to investigate the functionality of such overlay networks. However, much less attention has been paid to their reliability.
A common practice in conventional telecommunications systems using stored program control is to utilize redundant resources as the basis for a maintenance subsystem. In particular, central processors are duplicated and operated in synchronism so that fault detection can be accomplished by comparison methods. One such method uses software analysis programs to determine if a processor is faulty, so the faultless processor can be used to take over the load. A characteristic of this method is that the process of switching control over to one of the two processors affects all of the data needed for a communication path. In other words, the control system (or switching control system or processor control system) is potentially capable of handling any new call setup request while the faulty processor is being repaired. By multiplying the switching resources in a telecommunications network, the maintenance subsystem can block the use of a faulty resource, but a plurality of identical resources will still remain available so that the effect of the faulty resource on the communications traffic will be negligible. Consequently, using redundancy methods, an extremely high degree of security can be achieved in conventional telecommunications systems that provide basic switching functions and subscriber services.
However, the recent introduction of advanced telecommunications services provided by distributed computers has raised a number of questions related to operational reliability. One distributed network architecture that is being used to introduce these advanced services is known as Intelligent Networks (IN). An IN architecture is typically characterized by its relationship with an existing underlying network, whereby the IN hardware and software are closely integrated with the existing underlying network, and one operator controls the operation and maintenance of the complete network including the IN extensions. Consequently, it is a common practice to use the same methods for maintaining a reliable operation throughout the extended network including the IN nodes.
A problem with the solutions used by operators of the conventional telecommunications networks in attempting to achieve very high system availability is that these solutions are complex and costly. Furthermore, an independent service provider may not have access to all of the capabilities required from a telecommunications network. Even though it is sometimes argued that a lower degree of reliability may be acceptable regarding the new services provided through overlayed networks, it is clear that system reliability will still be an important and competitive factor. As described below, a problem that the present invention resolves is how to design a method that is easy to implement in emerging service networks overlayed on conventional networks, which provides a high degree of security at a low cost. The present invention achieves this goal partly by relying on the very high degree of security of the basic services provided by the underlying network.